47 cards generated

在牌组消失前保存

这些记忆卡还没保存——离开页面后会消失。创建一个免费账户来保留它们，并解锁下面的所有功能。

保存并复习
  • Save this deck to your account
  • Study with spaced repetition
  • Export to Anki (.apkg) or PDF
更大更强的生成
  • Process documents up to 100 pages
  • Images extracted from your PDFs
  • Sharper text extraction & a more advanced AI model
Sign up free → Free forever · No credit card
Copy of Final Exam Study Guide
  • 信息
    点击 'Save & Edit' 以启用全部功能
    学习
    导出
    编辑
    管理
公开

Flashcards in this deck (47)

正在搜索...

  • What are the levels of measurement in statistics?

    • Nominal: Categorical (e.g., Gender)
    • Ordinal: Continuous (e.g., Likert scale)
    • Interval: Continuous (e.g., Temperature)
    • Ratio: Continuous (e.g., Height)
    statistics measurement

  • What is reported for Nominal measurement?

    • Mean
    • Frequency
    statistics nominal

  • What should be reported for Interval measurement?

    • Mean
    • Standard Deviation
    • Histogram
    • Range
    statistics interval

  • What is reported for Ordinal measurement?

    • Mean
    • Standard Deviation
    • Histogram
    • Range
    statistics ordinal

  • What should be reported for Ratio measurement?

    • Mean
    • Standard Deviation
    • Histogram
    • Range
    statistics ratio

  • How to determine if the DV is continuous or categorical?

    Identify if the dependent variable (DV) is measured on a continuous scale or in categories.

    statistics dv

  • What test is used for a Continuous DV and Categorical IV?

    ANOVA

    statistics anova

  • What tests are used for Continuous DV and Continuous IV?

    • Correlation
    • Regression
    statistics correlation regression

  • Which test applies to Continuous DV and Repeated Categorical IV?

    Repeated Measures ANOVA

    statistics rm-anova

  • What test is used for Categorical DV and Categorical IV?

    Chi-Square

    statistics chi-square

  • What test is used for Categorical DV and Continuous or Categorical IV?

    Logistic Regression

    statistics logistic

  • What are the key assumptions for ANOVA?

    • Normality
    • Homogeneity of variance
    • Independence
    statistics anova assumptions

  • What additional assumption does Repeated Measures ANOVA require?

    Sphericity

    statistics rm-anova assumptions

  • What are the key assumptions for Correlation tests?

    • Linearity
    • No major outliers
    • Continuous variables
    statistics correlation assumptions

  • What assumptions are needed for Linear Regression?

    • Linearity
    • Homoscedasticity
    • Normal residuals
    • Independence
    statistics regression assumptions

  • What are the assumptions for Multiple/Hierarchical Regression?

    Same as Linear Regression

    statistics multiple assumptions

  • What is a key assumption for Regression?

    Expected cell counts ≥ 5 and independence.

    statistics regression

  • List a key assumption for Chi-square tests.

    Linearity of logit for continuous predictors; no multicollinearity.

    statistics chi-square

  • What is a primary condition for Logistic Regression?

    No linear assumptions; needs enough sample size; prone to overfitting.

    statistics logistic_regression

  • Outline a requirement for CART models.

    No linear assumptions; needs enough sample size; prone to overfitting.

    statistics cart

  • When should you report the Mean?

    If the variable is continuous (nominal); along with frequency.

    statistics descriptive

  • Under which condition should you report the Median?

    If there are outliers.

    statistics descriptive

  • When is the Mode reported?

    If categorical (interval, ratio, ordinal); including range and standard deviation.

    statistics descriptive

  • Which test is used when you have 1 continuous DV and 1 categorical IV?

    ANOVA

    Chi-Square

    Correlation

    Logistic regression

    statistics inferential

  • What type of ANOVA involves repeated measures on participants?

    Logistic ANOVA

    One-Way ANOVA

    RM ANOVA

    Two-Way ANOVA

    statistics anova

  • When should Linear regression be applied?

    2 variable DV + 1 continuous IV

    2 categorical DV

    1 cat DV + 1 continuous IV

    Correlation

    statistics regression

  • What assumptions are checked for between subjects design?

    Multicollinearity, Linearity, Lack of overfitting

    Normality, Homogeneity of variance, Independence of Observations

    statistics anova

  • Why is random assignment to treatments important?

    It helps ensure that treatment groups are equivalent in all respects, reducing bias.

    statistics experiments

  • What do we check for in RM ANOVA?

    • Normality (Shapiro-Wilk)
    • Sphericity (Mauchly's test)
    statistics anova

  • When are Post Hoc Tests conducted?

    Only if ANOVA is significant.

    statistics anova

  • What does Tukey HSD do?

    It conducts pairwise comparisons between conditions.

    statistics anova

  • What is reported after a Shapiro-Wilk test?

    State normal distribution or not with variable= W(x)=, p=x.

    statistics normality

  • What does a Levene's test assess?

    Homogeneity of variance.

    statistics anova

  • What is the implication of a significant one-way ANOVA?

    There is a significant relationship between x & y.

    statistics anova

  • What is the role of the Mauchly's test?

    It checks for sphericity in repeated measures ANOVA.

    statistics anova

  • What is a Bivariate Correlation?

    Measures the direction and degree of the linear relationship between two continuous variables.

    statistics correlation

  • When is Spearman Correlation used?

    For ordinal variables.

    statistics correlation

  • What does linear regression analyze?

    The relationship between one dependent variable and one or more independent variables.

    statistics regression

  • What are key concepts in analyzing research data?

    • Moderation models
    • Logistic regression
    • Descriptive statistics
    • Inferential statistics test selection
    • Assumptions for ANOVA and regression tests
    statistics research

  • What does a change in one variable indicate in a linear relationship?

    There is a constant change in the second variable.

    statistics linear_regression

  • What are the assumptions for linear regression?

    • Interval or ratio measurement
    • Related pairs
    • Absence of outliers
    • Linearity
    • Homoscedasticity
    statistics regression

  • What does the presence of outliers affect in regression analysis?

    They violate the assumption of linear regression.

    statistics outliers

  • What does a significant positive correlation indicate?

    X is related to Y; example: r(x)=0.384, p<0.05.

    statistics correlation

  • What is the equation of a simple linear regression line?

    y = bX + a

    statistics regression

  • When should multiple regression be used?

    When there are multiple predictors and one continuous outcome.

    statistics multiple_regression

  • List the assumptions needed for multiple regression.

    • Normality of residuals
    • Homoscedasticity
    • Linearity
    • Absence of multicollinearity
    statistics assumptions

  • What indicates multicollinearity in multiple regression?

    Tolerance < 0.2 and VIF > 10.

    statistics multicollinearity
学习笔记

Exam Overview

  • Focus on moderation models, chi-square, logistic regression, and CART for open-ended questions.
  • Review material from the midterm exam.

1. Variables and Descriptive Statistics

a. Levels of Measurement

  • Nominal: Categorical (e.g., Gender, Color); report Mean and Frequency.
  • Interval: Continuous (no true 0, e.g., Temperature, IQ); report Mean, SD, Histogram, Range.
  • Ordinal: Continuous (e.g., Likert scale); report Mean, SD, Histogram, Range.
  • Ratio: Continuous (true 0, e.g., Height, Weight); report Mean, SD, Histogram, Range.

b. Identifying DV Type

  • Determine if the dependent variable (DV) is continuous or categorical.

c. Choosing the Right Test

DV Type IV Type Test
Continuous Categorical ANOVA
Continuous Continuous Correlation or Regression
Continuous Categorical (repeated) RM-ANOVA
Categorical Categorical Chi-Square
Categorical Cont or Cat Logistic Regression
Test Key Assumptions
ANOVA Normality, Homogeneity of Variance, Independence
RM-ANOVA Plus Sphericity
Correlation Linearity, No Major Outliers
Linear Regression Linearity, Homoscedasticity, Normal Residuals
Multiple/Hierarchical Same as above

2. Inferential Statistics

  • ANOVA: 1 continuous DV + 1 categorical IV
  • RM ANOVA: repeated measures on participants
  • Linear Regression: 2 variable DV + 1 continuous IV
  • Logistic Regression: Continuous/Categorical IV + Dichotomous DV
  • Correlation: 2 continuous variables
  • Chi-Square: 2 categorical
  • Hierarchical Linear Regression: with multiple predictors
  • Moderation: ANOVA & multiple regression with 1 Cont DV + 2 Cat IVs
  • CART: Nonlinear/tree-based relationships

3. One-Way ANOVA

a. Between Subjects

  • Test differences across distinct participant groups.

b. Repeated Measures

  • Test differences using the same participants across conditions.

c. Assumptions

  1. Normality: Each group has a normal distribution of DV.
  2. Homogeneity of Variance: Same variance across groups; check via Levene's test.
  3. Independence of Observations: DV scores are unrelated.

4. RM ANOVA

Assumptions

  1. Normality: Use Shapiro-Wilk test.
  2. Sphericity: Check with Mauchly's test.

b. Post Hoc Tests

  • Conducted only if ANOVA is significant; explore mean differences.
  • Example: Tukey HSD.

5. Bivariate Correlation and Linear Regression

Correlation

  • Measures degree of linear relationship between two continuous variables.
  • Spearman: for ordinal variables.

Linear Regression

  • Makes predictions based on changes in one variable.
  • Assumptions:
  • Level of measurement: Interval or Ratio.
  • Related pairs.
  • Absence of outliers.
  • Linearity.
  • Homoscedasticity: constant variance around regression line.

6. Multiple Regression

When to Use

  • Multiple continuous or categorical predictors with one continuous outcome.

Assumptions

  1. Normality of Residuals: Should be normally distributed.
  2. Homoscedasticity: Residuals' variance is constant.
  3. Linearity: Linear association between predictors and outcome.
  4. Absence of Multicollinearity: Predictors shouldn't be highly correlated.